How do you set data quality policies?

In data governance, assessing quality needs isn't a one-time task. Business goals and technology evolve; your data quality strategy should too. That's why I advocate for quarterly 'Pulse Checks.' A good data governance leader or team should revisit their data sources, consumers, and key processes. Is everything aligned with your current objectives? Are there new variables you haven’t considered? Doing these checks adds a dynamic layer to your static data assessment. Your goal is not to just check a boxes. You really want to ensure that you're in tune with business needs. Adaptability is as vital as accuracy. So don't just assess your data quality needs—continuously reassess them.

ROI assessment on DQ engagements is something that I've seen missing in most of the Data Management Programs. DQ rules might get obsolete, lose their relevance over time or might not be tuned to Business KPIs. A proper dollar figure needs to be attached with each Business/Functional KPIs or usecases to understand the impact of bad data in terms of financial loss, only then we can do informed investment with tailored solutions in DQ programs which will bring an allover positive ROI to the organization from an Enterprise Data Management standpoint

It is imperative to understand and introspect the current state of data before implementing a DQ program. There may still be some DQ happening in the current processes and systems, but whether it is sufficient and relevant to business context, brings value and aligns to the organisational goals and objectives is the point of concern here. Data Quality is an integral part of the Data Governance program and should align with the strategic vision.

Absolutely, the strategic importance of clearly defining and prioritizing data quality dimensions cannot be overstated. By setting specific targets and thresholds, organizations not only ensure the reliability of their data but also align their data management practices with their strategic objectives.

Identify relevant dimensions top-down: from business strategy, through process KPIs, to the most critical data defects. This approach ensures that each data quality dimension directly supports your business objectives. For transparency, it is crucial to have clear methodologies and definitions that everyone understands. This clarity allows for effective communication and alignment across the organization, ensuring that everyone works towards the same data quality targets. It is not just about having policies, it is about having policies that are lived and breathed by the entire team.

Clearly state the objectives of your data quality policies. What do you want to achieve in terms of data accuracy, completeness, consistency, and timeliness?

Todo dado deve passar por algum tipo de chancela, isso trás mais confian?a para todo setor de Analytics, hoje praticamos uma avalia??o em duas camadas, negócios e técnica e esse alinhamento gera resultados incríveis. Podemos até nos basear em modelos pré existentes, mas somente alinha com nossas necessidades internas chegaremos ao patamar ideal de qualidade.

Establishing data roles, defining their responsibilities, and their interplay is crucial. Especially, the concept of ownership for data and its quality is key: thus, the role of a "data owner" is mission-critical. While roles are important, their introduction needs to be handled with care: at best, there is no new position established but rather existing structures, roles, and boards are enhanced with data quality responsibilities. This "non-invasive" approache ensures a lean data organization, which is closely intertwined with the existing organization - leading to a high understanding of the business needs and a quick roll-out.

I don’t think it is necessary to have a separate policy on data quality. Data quality can be covered by an umbrella data governance policy just fine. From my experience, it is important that, through established communication lines for data governance and constant work on raising data literacy and awareness, dedicated data experts from different teams work on definitions and rules. These should then be regularly monitored, and issues effectively addressed, strongly supported by the management of the organization.

Aqui está um outro ponto crucial quando falamos de dados em geral, n?o somente sobre qualidade. Precisamos entender os níveis que temos em nossa corpora??o e comunicar de forma adequada com cada uma dela. Criar um modelo de maturidade em dados ajuda nesse processo e pode nos ajudar a entender como evoluir passo a passo em cada unidade de negócio.

Deploying tools like data quality software and dashboards is just as important as ensuring all stakeholders are well-trained and aligned with these policies. The ADKAR model is a fantastic framework for managing this change effectively, focusing on awareness, desire, knowledge, ability, and reinforcement. Creating a culture where data quality is valued, understood, and actively pursued by everyone in the organization is crucial. Excellent approach to making data quality a cornerstone of organizational practice! #DataQuality #ChangeManagement #ADKAR #DataGovernance

Data Quality Policies Data quality policies encompass a comprehensive set of guidelines, procedures, and standards established by an organization to ensure the accuracy, completeness, reliability, and consistency of its data. These policies are pivotal in upholding the integrity of data and facilitating well-informed decision-making. Here are the key components typically included in data quality policies: 1. Data Definition and Classification: ? Clearly define what constitutes data for the organization. ? Classify data based on sensitivity, criticality, and purpose, specifying which components are Personally Identifiable Information (PII) and critical.

One should review and update data quality policies when the organization's business or data landscape changes, or when stakeholders raise concerns. Review annually or biannually to ensure policies are up-to-date.

2. Data Ownership and Stewardship: ? Explicitly assign responsibilities for data quality. ? Define roles such as data owners (responsible for the accuracy and integrity of specific data sets) and data stewards (responsible for overseeing the overall quality of data within the organization). 3. Data Collection and Entry Standards: ? Establish standards for how data is collected, entered, and recorded. ? This may include guidelines for data formats, validation rules, and data entry procedures. It's important to strike a balance to maintain validation rules without unduly slowing down the process.

1. Document and Socialize your Data policy with both business and technical stakeholders 2. Incorporate the policy in Architecture 3. Enforce the policy as code in Implementation 4. Monitor /Audit policy in a given period of time